I was attending an Intel presentation last week on their cloud strategy and what stuck me was the amount of internal alignment that had to be completed in order for them to take advantage of cloud based services. Now most organizations are not the size of Intel which runs 100000 servers across 95 datacenters but there are lessons to be learned from their cloud readiness preparation that are applicable to organizations of all sizes. The areas of focus were compute consistency which when we look at consuming Infrastructure as a Service (IaaS) relates to ensuring standard sizing of a virtual machine to an application workload. In our virtualization practice we recommend that customers standardize on a set of configurations for high, medium and low workloads if they do not include performance measuring as an internal process. This ensures consistency when transferring on premise workloads to IaaS providers. For companies running large virtualization shops this can be daunting if no consistent standards existed in the first place. Ironically Amazon just announced micro instances on EC2 for smaller application workloads which highlights the need for this type of categorization and standard http://aws.amazon.com/ec2/ .
The second area of focus was on consistent service standards. This is somewhat intuitive but may not be at the top of the list for customers considering consuming resources from the cloud. It stands to reason that if you are going to move compute resources to the cloud that you need to ensure that your services are well defined and consistent so either they can be matched by the provider or seamlessly maintained by your internal IT organization irrespective of geography. Really the message is rather simple but it is important; ensure you IT house is in order so consistency can be maintained when you take advantage of cloud based services.
- Posted using BlogPress from my iPad